coursera_deep_neural_network/7.1_simple1hiddenlayer.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<center>\n",
    "    <img src=\"https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBMDeveloperSkillsNetwork-DL0110EN-SkillsNetwork/Template/module%201/images/IDSNlogo.png\" width=\"300\" alt=\"cognitiveclass.ai logo\"  />\n",
    "</center>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1>Simple One Hidden Layer Neural Network</h1>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2>Objective</h2><ul><li> How to create simple Neural Network in pytorch.</li></ul> \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2>Table of Contents</h2>\n",
    "<p>In this lab, you will use a single-layer neural network to classify non linearly seprable data in 1-Ddatabase.</p>\n",
    "\n",
    "<ul>\n",
    "    <li><a href=\"#Model\">Neural Network Module and Training Function</a></li>\n",
    "    <li><a href=\"#Makeup_Data\">Make Some Data</a></li>\n",
    "    <li><a href=\"#Train\">Define the Neural Network, Criterion Function, Optimizer, and Train the Model</a></li>\n",
    "</ul>\n",
    "<p>Estimated Time Needed: <strong>25 min</strong></p>\n",
    "\n",
    "<hr>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2>Preparation</h2>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll need the following libraries\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:41:02.657551Z",
     "start_time": "2021-07-15T07:41:02.107487Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<torch._C.Generator at 0x7f35541d9770>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Import the libraries we need for this lab\n",
    "\n",
    "import torch \n",
    "import torch.nn as nn\n",
    "from torch import sigmoid\n",
    "import matplotlib.pylab as plt\n",
    "import numpy as np\n",
    "torch.manual_seed(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Used for plotting the model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:41:03.757094Z",
     "start_time": "2021-07-15T07:41:03.751206Z"
    }
   },
   "outputs": [],
   "source": [
    "# The function for plotting the model\n",
    "\n",
    "def PlotStuff(X, Y, model, epoch, leg=True):\n",
    "    \n",
    "    plt.plot(X.numpy(), model(X).detach().numpy(), label=('epoch ' + str(epoch)))\n",
    "    plt.plot(X.numpy(), Y.numpy(), 'r')\n",
    "    plt.xlabel('x')\n",
    "    if leg == True:\n",
    "        plt.legend()\n",
    "    else:\n",
    "        pass"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<!--Empty Space for separating topics-->\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2 id=\"Model\">Neural Network Module and Training Function</h2> \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define the activations and the output of the first linear layer as an attribute. Note that this is not good practice. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:41:08.827163Z",
     "start_time": "2021-07-15T07:41:08.821811Z"
    }
   },
   "outputs": [],
   "source": [
    "# Define the class Net\n",
    "\n",
    "class Net(nn.Module):\n",
    "    \n",
    "    # Constructor\n",
    "    def __init__(self, D_in, H, D_out):\n",
    "        super(Net, self).__init__()\n",
    "        # hidden layer \n",
    "        self.linear1 = nn.Linear(D_in, H)\n",
    "        self.linear2 = nn.Linear(H, D_out)\n",
    "        # Define the first linear layer as an attribute, this is not good practice\n",
    "        self.a1 = None\n",
    "        self.l1 = None\n",
    "        self.l2=None\n",
    "    \n",
    "    # Prediction\n",
    "    def forward(self, x):\n",
    "        self.l1 = self.linear1(x)\n",
    "        self.a1 = sigmoid(self.l1)\n",
    "        self.l2=self.linear2(self.a1)\n",
    "        yhat = sigmoid(self.linear2(self.a1))\n",
    "        return yhat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define the training function:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:41:15.617721Z",
     "start_time": "2021-07-15T07:41:15.613421Z"
    }
   },
   "outputs": [],
   "source": [
    "# Define the training function\n",
    "\n",
    "def train(Y, X, model, optimizer, criterion, epochs=1000):\n",
    "    cost = []\n",
    "    total=0\n",
    "    for epoch in range(epochs):\n",
    "        total=0\n",
    "        for y, x in zip(Y, X):\n",
    "            yhat = model(x)\n",
    "            loss = criterion(yhat, y)\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "            optimizer.zero_grad()\n",
    "            #cumulative loss \n",
    "            total+=loss.item() \n",
    "        cost.append(total)\n",
    "        if epoch % 300 == 0:    \n",
    "            PlotStuff(X, Y, model, epoch, leg=True)\n",
    "            plt.show()\n",
    "            model(X)\n",
    "            plt.scatter(model.a1.detach().numpy()[:, 0], model.a1.detach().numpy()[:, 1], c=Y.numpy().reshape(-1))\n",
    "            plt.title('activations')\n",
    "            plt.show()\n",
    "    return cost"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<!--Empty Space for separating topics-->\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2 id=\"Makeup_Data\">Make Some Data</h2>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:41:23.232128Z",
     "start_time": "2021-07-15T07:41:23.227892Z"
    }
   },
   "outputs": [],
   "source": [
    "# Make some data\n",
    "\n",
    "X = torch.arange(-20, 20, 1).view(-1, 1).type(torch.FloatTensor)\n",
    "Y = torch.zeros(X.shape[0])\n",
    "Y[(X[:, 0] > -4) & (X[:, 0] < 4)] = 1.0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<!--Empty Space for separating topics-->\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2 id=\"Train\">Define the Neural Network, Criterion Function, Optimizer and Train the Model</h2>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create the Cross-Entropy loss function: \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:41:25.519988Z",
     "start_time": "2021-07-15T07:41:25.517491Z"
    }
   },
   "outputs": [],
   "source": [
    "# The loss function\n",
    "\n",
    "def criterion_cross(outputs, labels):\n",
    "    out = -1 * torch.mean(labels * torch.log(outputs) + (1 - labels) * torch.log(1 - outputs))\n",
    "    return out"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define the Neural Network, Optimizer, and Train the Model:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:41:48.603122Z",
     "start_time": "2021-07-15T07:41:32.153411Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'cross entropy loss')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Train the model\n",
    "# size of input \n",
    "D_in = 1\n",
    "# size of hidden layer \n",
    "H = 2\n",
    "# number of outputs \n",
    "D_out = 1\n",
    "# learning rate \n",
    "learning_rate = 0.1\n",
    "# create the model \n",
    "model = Net(D_in, H, D_out)\n",
    "#optimizer \n",
    "optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)\n",
    "#train the model usein\n",
    "cost_cross = train(Y, X, model, optimizer, criterion_cross, epochs=1000)\n",
    "#plot the loss\n",
    "plt.plot(cost_cross)\n",
    "plt.xlabel('epoch')\n",
    "plt.title('cross entropy loss')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By examining the output of the  activation, you see by the 600th epoch that the data has been mapped to a linearly separable space.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "we can make a prediction for a arbitrary one tensors \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:44:19.010395Z",
     "start_time": "2021-07-15T07:44:19.006616Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([0.9969], grad_fn=<SigmoidBackward>)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x=torch.tensor([0.0])\n",
    "yhat=model(x)\n",
    "yhat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "we can make a prediction for some arbitrary one tensors  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:44:22.183537Z",
     "start_time": "2021-07-15T07:44:22.176606Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0.9969],\n",
       "        [0.9963],\n",
       "        [0.9680]], grad_fn=<SigmoidBackward>)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_=torch.tensor([[0.0],[2.0],[3.0]])\n",
    "Yhat=model(X_)\n",
    "Yhat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "we  can threshold the predication\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:44:23.919952Z",
     "start_time": "2021-07-15T07:44:23.909294Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[True],\n",
       "        [True],\n",
       "        [True]])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Yhat=Yhat>0.5\n",
    "Yhat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3>Practice</h3>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Repeat the previous steps above by using the MSE cost or total loss: \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-07-15T07:46:52.870767Z",
     "start_time": "2021-07-15T07:46:38.966194Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/explore/miniconda3/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/loss.py:446: UserWarning: Using a target size (torch.Size([])) that is different to the input size (torch.Size([1])). This will likely lead to incorrect results due to broadcasting. Please ensure they have the same size.\n",
      "  return F.mse_loss(input, target, reduction=self.reduction)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/explore/miniconda3/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/loss.py:446: UserWarning: Using a target size (torch.Size([])) that is different to the input size (torch.Size([1])). This will likely lead to incorrect results due to broadcasting. Please ensure they have the same size.\n",
      "  return F.mse_loss(input, target, reduction=self.reduction)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEGCAYAAAB1iW6ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3de3Rc5X3u8e9vRqPLSLJsXXyVr2AbDDiEyEACJ3AKJMChEE5uZDmFppxAktKmaZsuekh8YkJWmiZputrQckhDCORGQk+Cm5pCSnDSXAi2A9gYY2wcG0vYWJJlWdLoMtK854+9xx7LI2k0GmtmzzyftbQ0s2dL+nkzenj17vdizjlERCT4QvkuQEREckOBLiJSJBToIiJFQoEuIlIkFOgiIkWiLF8/uLGx0S1ZsiRfP15EJJC2bt3a4ZxrSvda3gJ9yZIlbNmyJV8/XkQkkMxs/1ivqctFRKRIKNBFRIqEAl1EpEjkrQ9dRIIvHo/T2trKwMBAvkspOpWVlTQ3NxOJRDL+GgW6iGSttbWV2tpalixZgpnlu5yi4Zyjs7OT1tZWli5dmvHXTdjlYmYPmNlhM3txjNfNzP7BzPaY2TYzu2ASdYtIgA0MDNDQ0KAwzzEzo6GhYdJ/+WTSh/4gcPU4r18DLPc/bgP+eVIViEigKcxPj2yu64RdLs65n5vZknFOuQF4yHnr8D5jZjPNbJ5z7uCkqxHJxqZN8NOf5ruK7IRCcPPNsGxZviuRIpCLPvQFwIGU563+sVMC3cxuw2vFs2jRohz8aBHgr/4KNm+GILYUnYO+PvjiF/NdiaSxadMmvvSlL/HjH/943PM+/elP89hjjxEKhZg9ezYPPvgg8+fPxznHxz/+cTZu3Eg0GuXBBx/kggu8XulvfvOb3HPPPQB86lOf4pZbbplyvdM6bNE5d79zrsU519LUlHbmqsjk9fTA+94HiUTwPmbPht7efF9BmaJPfvKTbNu2jeeff57rrruOu+++G4DHH3+c3bt3s3v3bu6//34++tGPAnDkyBHWr1/Pb37zG5599lnWr19PV1fXlOvIRaC3AQtTnjf7x0SmRywG0Wi+q8hONOrVL1n71re+xYUXXsj555/P7bffzsjICAA1NTV84hOf4JxzzuGKK66gvb0dgOeff56LL76Y1atXc+ONNx4P0j179nDllVfypje9iQsuuIBXX30VgN7eXt7znvdw1llnsXbtWtLt8jZjxozjj/v6+o73fz/22GPcfPPNmBkXX3wxR48e5eDBgzzxxBNcddVV1NfXM2vWLK666ir+4z/+Y8rXIhddLhuAO8zse8BFQLf6z2VaKdALwvp/28FLrx/L6fdcNX8G/+f3zxnz9Z07d/LII4/wy1/+kkgkwsc+9jG+/e1vc/PNN9PX10dLSwtf+cpXuPvuu1m/fj1f/epXufnmm/nHf/xHLrvsMtatW8f69ev5+7//e9auXcudd97JjTfeyMDAAIlEggMHDvDcc8+xY8cO5s+fzyWXXMIvf/lLLr300lNqueuuu3jooYeoq6vj6aefBqCtrY2FC0+0d5ubm2lraxvz+FRlMmzxu8CvgZVm1mpmt5rZR8zsI/4pG4G9wB7ga8DHplyVyGQo0EvWU089xdatW1mzZg3nn38+Tz31FHv37gUgFArx/ve/H4APfvCD/OIXv6C7u5ujR49y2WWXAXDLLbfw85//nJ6eHtra2rjxxhsBb1JP1H9PXXjhhTQ3NxMKhTj//PPZt29f2lo+97nPceDAAdauXctXv/rV0/wvTy+TUS4fmOB1B/xxzioSmQznFOgFYryW9OninOOWW27h85///ITnZju8sqKi4vjjcDjM8PDwuOevXbuWa6+9lvXr17NgwQIOHDgxZqS1tZUFCxawYMECNm3adNLxyy+/PKv6UmktFwm25MQLBXpJuuKKK3j00Uc5fPgw4N1s3L/fW102kUjw6KOPAvCd73yHSy+9lLq6OmbNmsV//dd/AfDwww9z2WWXUVtbS3NzMz/60Y8AGBwcJDaJ/y67d+8+/vixxx7jrLPOAuD666/noYcewjnHM888Q11dHfPmzeOd73wnTz75JF1dXXR1dfHkk0/yzne+c8rXQ1P/JdiSv3RBDvTW1nxXEVirVq3innvu4R3veAeJRIJIJMK9997L4sWLqa6u5tlnn+Wee+5h9uzZPPLII4A3XPAjH/kIsViMZcuW8Y1vfAPwwv32229n3bp1RCIRfvCDH2Rcx5133smuXbsIhUIsXryY++67D4Brr72WjRs3cuaZZxKNRo//rPr6ej796U+zZs0aANatW0d9ff2Ur4elu2M7HVpaWpw2uJApO3AAFi2Cf/kXuPXWfFczeX/wB/CrX4E/oiJodu7cydlnn53vMtKqqamhN+BDQtNdXzPb6pxrSXe+ulwk2Iqhha4uF8kRBboEW1+f97m6Or91ZKu6+sS/QXIq6K3zbCjQJdiKpYWep67PXMhXt22xy+a6KtAl2Ioh0EdGIB7PdyVZqayspLOzU6GeY8n10CsrKyf1dRrlIsGWw0B3znG4Z5D9nTFeOxLjtc4+9h+JcTQWZ+XcWlY31/Gm5pk0z6rK3ZKxybpjMSgvz833nEbNzc20trYen1YvuZPcsWgyFOgSbFMM9Laj/Xx+4052HerhQFeMgXji+Gshg/kzq6itjPDrVzsZGvFeq68u57wFdaxurmN180zWLJnFzGiWYZwa6DNnZvc98igSiUxqRx05vRToEmxTCPQXDhzl1m9uYSA+wlvPaOCyFU0sboiysD7K4oZqFsysorzM65UcGk6w61APL7QeZVvrUba1dnPv0+0kHMyMRvjXj76NM5pqJl9/aqCLTJECXYIty0B/fPtBPvH952msqeC7H76I5XNqxz2/vCzEec11nNdcBywGoH9ohOcOdPEn33mOD31jMz/82NtoqKkY9/ucQoEuOaSbohJskwx05xz3/exVPvrt33L2vBn86I8vmTDMx1JVHuZtZzTytVtaeOPYALc9vJWB+MjkvokCXXJIgS7BFot5OxVVTNwyHhpOcOe/budvHn+Z61bP47sfvpjGybao07hg0Sy+8v7z2bq/i7/8wQskEpMY8aFAlxxSoEuw9fV5oTjBqJPuWJw//MazPLLlAH/6e2fyDze9mcpIOGdlXHvePO685ix+vO0gX/7Jrsy/MBnomlwkOaA+dAm2WGzCWaL7O/v40IObOXAkxpff+ybe/ZbJDQXL1O1vX8b+zj7uffpVFtdX8741Cyf+omTtaqFLDijQJdgmWAt9JOG46f5n6I+P8K1bL+KiZQ2nrRQz4+4bzqW1q5///cPtLJhVxSVnNo7/RepykRxSl4sE2wSB/vKhYxzsHmDddatOa5gnRcIh7l17AWc01fCRb21l9xs943+BAl1ySIEuwTZBoG/d720AvGbJ1NeaztSMyggPfGgNlZEwH3pwM+09g2OfrECXHFKgS7BNEOhb9nUxZ0YFzbOqprEoWDCziq/f0kJH7yCf2bBj7BOr/LoU6JIDCnQJtgxa6C2L63O39sokrG6eyTXnzuO3r3WNfVIoBJWVCnTJCQW6BNs4gf760X7ajvbzlsWzprmoE1bMqeVg9wDd/eOspqhNLiRHFOgSbOME+pY89J+PtmKOt77LuDdHFeiSIwp0CbZxAn3rviNEy8OcPS+7qf25sMJfVmCXAl2mgQJdgi05UzSNLfu7OH/hTMrC+XubL5hZRXV5mN1vjLMdWjSqmaKSEwp0CS7nxpwp2js4zM6Dx2jJY/85QChkLJ9Ty65D47TQq6vVQpecUKBLcMXj3vZtaVroz792lISDt+Sx/zxp5ZxaXlGXi0wDBboE1zhL527ZfwQzePOi/O8CtGJuLZ19Q3T0jjHBSIEuOaJAl+AaL9D3dXHW3BnMqIxMc1GnSo50eWWsbhcFuuSIAl2Ca4xAHx5J8NxrXXnvP09a6Y90GbPbRYEuOaJAl+AaI9BfPtRD39AILUsKI9CbaiuYGY2wa6yRLgp0yREFugTXGIGeXJArnzNEU5kZK8a7MapAlxzJKNDN7Goz22Vme8zszjSvLzKzp83sOTPbZmbX5r5UkVHGCPQt+7uYO6OSBTOnd0Gu8aycU8srh3pwLs32dNEoDA56I3ZEpmDCQDezMHAvcA2wCviAma0addqngO87594M3AT8U64LFTlFcjLO6Bb6viO8ZcmsvCzINZYVc2vpGRzmYPfAqS9qCV3JkUxa6BcCe5xze51zQ8D3gBtGneOAGf7jOuD13JUoMoZkAKZMLGo72s/r3QOsKZDulqQVs72RLmmXANA2dJIjmQT6AuBAyvNW/1iqzwAfNLNWYCPwJ+m+kZndZmZbzGxLe3t7FuWKpEjT5bJl3xEAWgpgQlGq5JouaRfpUgtdciRXN0U/ADzonGsGrgUeNrNTvrdz7n7nXItzrqWpqSlHP1pKVppA37q/i2h5mLPm5m9BrnRmVZczu7aCXYfSjHRRoEuOZBLobUDq9uXN/rFUtwLfB3DO/RqoBCbYHVdkitK20Lt486L8Lsg1lpVzxxjpokCXHMnkXb8ZWG5mS82sHO+m54ZR57wGXAFgZmfjBbr6VOT0Sgagv41b7+AwLx86xlsWF1Z3S9KKObXsPtzDSGLUSBcFuuTIhIHunBsG7gCeAHbijWbZYWZ3m9n1/ml/AXzYzF4Avgv8oUs7Pkskh2Ixb/u2kPc2fu61LhKOgpkhOtqKOTUMxBMcODIquBXokiNlmZzknNuId7Mz9di6lMcvAZfktjSRCYza3GLzvi5CBbIgVzorUpYAWNKYsuSvAl1ypPA6GkUyNSrQt+4/wllzZ1BbAAtypbN8rDVdFOiSIwp0Ca6U3Yq8BbmOFsz6LenUVJTRPKvq1DVdkoGuXYtkihToElwpLfSXD/UQGxopmPVbxpJcAuAkaqFLjijQJbhStp8r1AlFo62YW8vejl7iI4kTBxXokiMKdAmulBb6lv1dzKsrrAW50lkxp4b4iGNfR0r3SiTifSjQZYoU6BJcfqA759iyr6vgW+dwYqTLKWu6aAldyQEFugSXH+htR/s5dGygYMefpzqjqYaQpdmOToEuOaBAl+DyA73QNrQYT2UkzJLGarXQ5bRQoEtw+YG+ZV8X1QW4INdYVs6p5ZV0QxcV6DJFCnQJLj/Qt7d1s7q5MBfkSmfFnFr2d/YxEE/ZoUiBLjkQjN8AkdFGRrxt26JRDh8bYH6Bj25JtWJOLQkHew6ntNKjUU0skilToEsw+a1ZV1VFR+8QjbXleS4ocyvnersXnbQEgFrokgMKdAkmP/wGyqsYGknQVFOR54Iyt7ihmvJw6OQbo9XVCnSZMgW6BJMffj0hbyGuxgAFeiQcYllT9clDF9VClxxQoEsw+eHXbcELdEjuXjSqD12BLlOkQJdg8sPvKH6gB6gPHbwbo21H++kZiHsHFOiSAwp0CSY//I44b4+WoLXQk0sA7E6OdEkGujb6kilQoEsw+YHe4cKEDGZFg9VCX5nc7CLZj55ccXFgIE8VSTFQoEsw+YHePhKmvrqCcMjyXNDkNM+qoioSPjHSRUvoSg4o0CWY/Ek4bwyHaawJVuscIBQyVsypOTEWXbsWSQ4o0CWY/JbsobjRVBus/vOkFalruqiFLjmgQJdg8oPv9bgF7oZo0oo5tbT3DHKkb0iBLjmhQJdg8oOvbdAC2eUC3nZ04C8B4G+lp0CXqVCgSzDFYrhIhN5EKLAt9OMjXd7oUQtdckKBLsEUi+GqvBAMaqDPmVFBdXmYve19CnTJCQW6BFMsxnBlJQCNAb0paubd0O3oHVSgS04o0CWYYjHiFd4a6EHtQwcU6JJTCnQJpliMoXKvhR6kpXNHa6ypoKNXo1wkNxToEkx9fQxEKjCD+upgt9DbewY1sUhyQoEuwRSLESsrZ1a0PDB7iabTWFNBd3+cwXAZmKmFLlOS0W+CmV1tZrvMbI+Z3TnGOe8zs5fMbIeZfSe3ZYqMEovRF64IdP85nBih09kX1xK6MmVlE51gZmHgXuAqoBXYbGYbnHMvpZyzHPhr4BLnXJeZzT5dBYsAEIvRG50Z2CGLScllCzp6B5mvQJcpyqSFfiGwxzm31zk3BHwPuGHUOR8G7nXOdQE45w7ntkyRUWIxukORwAd68i+Mjt5B7SsqU5ZJoC8ADqQ8b/WPpVoBrDCzX5rZM2Z2dbpvZGa3mdkWM9vS3t6eXcUiALEYRymGQPfqP35jVIEuU5Cru0llwHLgcuADwNfMbObok5xz9zvnWpxzLU1NTTn60VKKXCzGsVB54LaeG+1El8uQAl2mLJNAbwMWpjxv9o+lagU2OOfizrnfAa/gBbxI7jmHxWIMlFUEvoVeGQlTW1GmFrrkRCaBvhlYbmZLzawcuAnYMOqcH+G1zjGzRrwumL05rFPkBH+btv5IRaAnFSU11lbQ3qtAl6mbMNCdc8PAHcATwE7g+865HWZ2t5ld75/2BNBpZi8BTwOfdM51nq6ipcT5odcfCX4LHbyZrh1qoUsOTDhsEcA5txHYOOrYupTHDvhz/0Pk9PJnU/aXVQS+Dx2gsbacXYf8JXQ1U1SmILhT7KR0+a3YgUgFDdXF0UJXH7rkggJdgscPPauOUl4W/LdwY00FxwaGveWAFegyBcH/bZDS44de+YzaPBeSG8n13GMRP9Cdy3NFElQKdAkeP9Ar64oj0JMjdfrC5TAyAvF4niuSoFKgS/D4gR6dWRyBnmyh94T8G7zqdpEsKdAlePzAq5k1I8+F5EZyPZduIt4BBbpkSYEugRPv6QWgpr5YAt1roXclRxEr0CVLCnQJnL6uYwDU1dfluZLcqIyEqa0sU6DLlCnQJXBifqDPbCyOQAd/s+gRP9A1uUiypECXwBno7iWBUd9QHF0u4HW7HE74v45qoUuWFOgSOEM9Pd46LrXBnyWa1FRTwRvxsPdEgS5ZUqBL4MR7+opmYa6kptoKDg6b90SBLllSoEvgjPT2MVBeSWUknO9ScqaxppyOhIYtytQo0CVwEn19DFdU5ruMnGqsqaA/4v/FoUCXLCnQJXhiMYYrq/JdRU411VbQX6ZAl6lRoEvghPr7SVRG811GTjXWVDAQ0dR/mRoFugROeKAfosXVQm+srcBZiOGKCgW6ZE2BLoEyNJygfLCfUHV1vkvJqeR6LvHyKk0skqwp0CVQOvsGqYoPEq4prkCvKAszo7KMoXK10CV7CnQJlI6eIaqGBymrLa5AB+/G6EBEgS7ZU6BLoHT0DlIZH6R8Rk2+S8m5xpoKYmUKdMleWb4LEJmM9p4BquKDuCLZfi5VY22Ft2uRAl2ypBa6BMqRo72UuQRVRbK5RaqmmgqOhdVCl+wp0CVQuju8pXMjtcXX5dJUW0FvuJxEnwJdsqNAl0Dp8ddCJ1pcE4vAG7rYX1bBiIYtSpYU6BIofV3d3oMiDPSmWm89F6dAlywp0CVQ+ou6he4FuqkPXbKkQJdAGej2Nogu1kAfKKsg1N+f71IkoBToEhjDIwnivcUb6A015fRHKgjHh2B4ON/lSAAp0CUwjvQNURkf9J4UYaBXlIVxVf6iY2qlSxYU6BIY7b3eOi5AUQY6pAzHVD+6ZCGjQDezq81sl5ntMbM7xznv3WbmzKwldyWKeDp6h04EepGttphUNkOBLtmbMNDNLAzcC1wDrAI+YGar0pxXC3wc+E2uixQB6OgZpHK4uFvoFQp0mYJMWugXAnucc3udc0PA94Ab0pz3WeALwEAO6xM5rqMEulyq6vw1ahTokoVMAn0BcCDleat/7DgzuwBY6Jz79/G+kZndZmZbzGxLe3v7pIuV0tbRO0htYsh7UlVcOxYlRWd6a9QMdvfkuRIJoinfFDWzEPB3wF9MdK5z7n7nXItzrqWpqWmqP1pKTEfvEPXEobISQsV5P7/GX3TsWOfRPFciQZTJb0UbsDDlebN/LKkWOBfYZGb7gIuBDboxKrnW0TvITIaLtrsFYEaDF+g9R47luRIJokwCfTOw3MyWmlk5cBOwIfmic67bOdfonFvinFsCPANc75zbcloqlpLV3jNInYsXdaDX1c8EoK9LgS6TN2GgO+eGgTuAJ4CdwPedczvM7G4zu/50FyiS1NE75PWhF3Ggz2qqA6BffeiShYx2LHLObQQ2jjq2boxzL596WSInG0k4jvQNUjNS7IHutdCPr1kjMgnFeWdJik5XbIiEg6rh4g705EzRoR4FukyeAl0CoaPXG39eFR8o2lmiAEQixMNlxHu0JrpMngJdAqGjxxt/Xj40WNQtdICh8kpGetVCl8lToEsgJFvokcH+og/04YpKnPYVlSwo0CUQkoFeNjBQ9IE+Ulmlqf+SFQW6BEJ77yDl4RAW6yv6QHfRKOWD/cSGtMmFTI4CXQKho2eIxppyb7/NIg90olGq4oPH7xuIZEqBLoHQ0TvI7GgYhop72CJAqDpK5fAg7X43k0imFOgSCB29g8wv958UeaCHa2qoig/S3qNAl8lRoEsgdPQOMjeS8J4UeaCX11Z7XS5qocskKdCl4CUSjs7eIeaU+YFezBOLgPK6WiqHFegyeQp0KXjd/XGGE46m8Ih3oMhb6KHqaqqHh9TlIpOmQJeCl2ypNoZKI9CJRqlSC12yoECXgpcc7VFvce9ACQR65dAAHWqhyyQp0KXgdfR647FnudIJdIBj2rVIJkmBLgUv2VKtc/7MyRIJ9L6jCnSZHAW6FLyO3kHKQkZ02O+CKJFAJxajb1DT/yVzCnQpeK1d/cyZUUlooN87UCKBrrHoMlkKdCl4L7Z1c+6CGSdWICyRQK9UoMskKdCloB0biLO3o4/VzTNLLtCrhjX9XyZHgS4F7cW2bgDOXVDnBXok4n0UM38mbFV8kPZerbgomVOgS0Hb3uoF+nnJQC/21jmc1ELXWHSZDAW6FLTtbd00z6qivrq85AK9MTSiJXRlUhToUtC2t3WzurnOe9JX/LsVAcf/jbNtWC10mRQFuhSs7lic/Z0xr/8cSq6FXh8a1igXmRQFuhSs7f4N0dULZnoHSizQZzGsLheZFAW6FKxtbUcB/4YolE6gV1SAGTOJ09EzhHMu3xVJQCjQpWC92NbN4oYodVF/mGKpBLoZRKPMcHH64yP0DY3kuyIJCAW6FKxtrd0n+s+hdAIdIBqldsQbg64bo5IpBboUpCN9Q7R29bO6hAM9mgx09aNLhjIKdDO72sx2mdkeM7szzet/bmYvmdk2M3vKzBbnvlQpJckbouc1jwr0It9P9LjqaqJxL8gPdg/kuRgJigkD3czCwL3ANcAq4ANmtmrUac8BLc651cCjwN/mulApLdtbvRuipdzlUp0YoioS5tnfHcl3NRIQmbTQLwT2OOf2OueGgO8BN6Se4Jx72jnnr5zEM0BzbsuUUrO9rZuljdXMqPRviCYSJRfo4ViMt53RwKZXDmuki2Qkk0BfABxIed7qHxvLrcDj6V4ws9vMbIuZbWlvb8+8Sik521u7TwxXBBjwux1KKNCJxbh8ZRMHjvSzt6Mv3xVJAOT0pqiZfRBoAb6Y7nXn3P3OuRbnXEtTU1Muf7QUkfaeQV7vHjgx5R9KZ+ncJD/QL1sxG4Cf7VIDSCaWSaC3AQtTnjf7x05iZlcCdwHXO+d0W16yllwy97zR/edQcoG+qCHKssZqNr2iQJeJZRLom4HlZrbUzMqBm4ANqSeY2ZuB/4sX5odzX6aUku1t3ZjBOQp0AC5b2cQzezvp1wQjmcCEge6cGwbuAJ4AdgLfd87tMLO7zex6/7QvAjXAD8zseTPbMMa3E5nQttZuljVWU1NRduJgCQf65StnMzSc4Jm9nXkuSgpd2cSngHNuI7Bx1LF1KY+vzHFdUsK2tx3lbWc0nnywVAPdOS5aWk9lJMSmXYf572fNzndlUsA0U1QKyhvHBnjj2ODJ/edwItBLaGIRIyMQj1MZCfPWZQ3qR5cJKdCloCS3nDtphAuUZgsdTvSjr2hif2eMfRq+KONQoEtB2d7WTchg1fwZJ7/Q5wdZqQW6/+++fKXX1bJpl8YcyNgU6FJQtrd1c+bsGqLlo27vlHgLfUljNUsaoup2kXEp0KVgOOfY1trNeckdilKVeKCD10r/9audDMQ1fFHSU6BLwTh0bICO3sFT+89BgY43Hn1QwxdlHAp0KRjJG6Lnjh7hAl6wmXnbs5WCNIH+1mUNVJSF+Jm6XWQMCnQpGNvbugmHjFXzZpz6YnKlRbPpLywf0gR6ZSTMRcsatK6LjEmBLgVjW2s3y2fXUFUePvXFUlo6F9IGOsDlK5rY29HHa52xNF8kpU6BLgXBOcf2tu70/eegQPddvtJbpXTTKxq+KKdSoEtBaDvaz5G+oVNniCaV0vZzcOLfOirQlzZWs6g+yiZ1u0gaCnQpCMeXzG1OM2QRvAk2pdhC7zt5ZqiZcfnKJg1flLQU6FIQtrV2UxYyzppbm/6EUutyqaryPsdO7Su/bEUT/fERNu/TXqNyMgW6FITtbd2snFtLZSTNDVEovUAPhaCyMm2gv/WMBsrDIXW7yCkU6JJ3yRmiY94QhdILdDhpTfSTDpeXcdGyeq3rIqdQoEvetXb1090fTz+hKEmBfpLLVjTxansfB45o+KKcoECXvHtixyEAVqdbwyVJgX6S46svataopFCgS159/Re/455/38lblzWcumRuKgX6Sc5oqmbBzCrNGpWTKNAlL5xz/M3jL/PZH7/E1efM5RsfWkM4NMa0fucU6KMkhy/+6tUOuvqGprkwKVQKdJl2wyMJPvnoNu772ausvWgR9669YOzRLQDxuLcdmwL9JO968wIGhxNc9ZWfseGF13HOTWNxUogU6DKt+odGuP3hrTy6tZU/u3I597zr3LFb5kmltp9oUnX1uIG+Zkk9G+64hPkzq/jT7z7HHz24mbaj/dNYoBQaBbpMm6OxIdb+yzM8vesw97zrXP7syhVYJqsnltr2c0nR6CkzRUc7Z34dP/zYJXzqf5zNM3uPcNXf/YwHfvE7RhJqrZciBbpMi9eP9vOe+37Ni23H+Ke1F/DBixdn/sWltrlF0gRdLknhkPG//tsynvzE21mzpJ67f/wS/+H6tO0AAAjjSURBVPOff8XLh45NQ5FSSMomPkVk8gbiI+w61MOO14+x4/VufvLSG/QPjfDQrRdy8bKGyX0zBXpGFtZHefBDa9jwwuus/7eXuO4ffsH71yzknPl1LGmIsrixmnkzKglN1MUlgaVAl0lxzjE0kiA2OELf0DCxoRH6BofpGRhmz+FeXny9m5deP8buw73H/+yvrSxjdXMdd127avyhiWNRoGfMzLjh/AW8fXkTn9u4kx9saeXbI68df728LMTCWVUsaahmcUM1TbUVVEZCVJSFqYyEqIyET3peFgoRMsMMQmaEQ0bIvJ+T/AxgnNh7xLCM9yEplf1KRpsZLaemIvfxG7hA/9FzbTz06335LiNrp7Nnc/Qgh1N+lnO4lPMcDudOPE84R8I5RhKOhIORhDvx4RyD8RFiQyMMj9M/21hTwbkLZnDF2bM5d34d58yvY2F9VWZ95WMp5UAfGoLhYSib3K/qrOpyvvTeN/GFd6/mYHc/+ztj/kcf+zr72N8Z41evdtKvFRvz4p53nTu5bscMBS7Qy8JG9Wn4P1uxGB2co2PULNmaGt2y8lpV4WQrLGSEDUIhoyzkHasoCxMtD1NdUXb8c3V5GdGKMNXlZSxpiDJ7RmXu/1GlHOgA/f1QO8YqlBMIh4zmWVGaZ0W55MyTX3POMRBPMBAfYWB4hMF4goHhkRPH4iPe/+QTJ/5nn3Cc+Oz/jz3ZMPC+54mGxETDKEv5tu0Fi2adlu8buGS8bvV8rls9P99lyHQq9UCPxbIO9PGYGVXl4fRb/kkgaZSLFD4Fen7rkMBQoEvhK+WJRaBAl4wp0KXwlfLEIphwcpFIUkaBbmZXm9kuM9tjZnemeb3CzB7xX/+NmS3JdaFSwpIt1OS2bKVCXS4ySRMGupmFgXuBa4BVwAfMbNWo024FupxzZwJfAb6Q60KlhMVi3nZsoRL7g1KBLpOUySiXC4E9zrm9AGb2PeAG4KWUc24APuM/fhT4qpmZOx3Lvz3wAHz5yzn/tlLADh4svdY5nAj022+HGVlMyJLCtW4dvP/9Of+2mQT6AuBAyvNW4KKxznHODZtZN9AAdKSeZGa3AbcBLFq0KLuKGxpg1eg/EKSorVoFF1+c7yqm31lneWHe2ZnvSiTXZhXBOHTn3P3A/QAtLS3Ztd5vuMH7ECl2kQjcd1++q5AAyaRTsg1YmPK82T+W9hwzKwPqADUrRESmUSaBvhlYbmZLzawcuAnYMOqcDcAt/uP3AD89Lf3nIiIypgm7XPw+8TuAJ4Aw8IBzboeZ3Q1scc5tAL4OPGxme4AjeKEvIiLTKKM+dOfcRmDjqGPrUh4PAO/NbWkiIjIZJTawV0SkeCnQRUSKhAJdRKRIKNBFRIqE5Wt0oZm1A/uz/PJGRs1CLSCqLTuqLTuqLTtBrm2xc64p3Qt5C/SpMLMtzrmWfNeRjmrLjmrLjmrLTrHWpi4XEZEioUAXESkSQQ30+/NdwDhUW3ZUW3ZUW3aKsrZA9qGLiMipgtpCFxGRURToIiJFIlCBbmZfNLOXzWybmf3QzGamvPbX/ibVu8zsnXmo7b1mtsPMEmbWknJ8iZn1m9nz/se071gwVm3+a3m9bqNq+YyZtaVcq2vzWY9f07gbpOeTme0zs+3+tdqS51oeMLPDZvZiyrF6M/uJme32P5+ebXqyq60g3mtmttDMnjazl/zf0Y/7x7O7ds65wHwA7wDK/MdfAL7gP14FvABUAEuBV4HwNNd2NrAS2AS0pBxfAryY5+s2Vm15v26j6vwM8Jf5fp+l1BP2r8kyoNy/VqvyXVdKffuAxnzX4dfyduCC1Pc68LfAnf7jO5O/rwVSW0G814B5wAX+41rgFf/3MqtrF6gWunPuSefcsP/0Gbzdk8DbpPp7zrlB59zvgD14m1tPZ207nXO7pvNnZmqc2vJ+3Qrc8Q3SnXNDQHKDdBnFOfdzvL0QUt0AfNN//E3gXdNalG+M2gqCc+6gc+63/uMeYCfeHs1ZXbtABfoofwQ87j9Ot5H1gmmvaGxLzew5M/uZmf23fBeTohCv2x1+l9oD+foTPUUhXp9UDnjSzLb6G7AXmjnOuYP+40PAnHwWk0YhvdcwsyXAm4HfkOW1m9ZNojNhZv8JzE3z0l3Oucf8c+4ChoFvF1ptaRwEFjnnOs3sLcCPzOwc59yxAqht2o1XJ/DPwGfxguqzwJfx/sct6V3qnGszs9nAT8zsZb81WnCcc87MCmmMdEG918ysBvhX4M+cc8fM7Phrk7l2BRfozrkrx3vdzP4QuA64wvkdTGS2kfVpr22MrxkEBv3HW83sVWAFkNObWNnUxjRdt1SZ1mlmXwN+fDprycC0X5/JcM61+Z8Pm9kP8bqICinQ3zCzec65g2Y2Dzic74KSnHNvJB/n+71mZhG8MP+2c+7/+YezunaB6nIxs6uBvwKud87FUl7aANxkZhVmthRYDjybjxpHM7MmMwv7j5fh1bY3v1UdV1DXzX/jJt0IvDjWudMkkw3S88LMqs2sNvkYb8BAvq/XaKmbx98CFNJfigXxXjOvKf51YKdz7u9SXsru2uX7Lu8k7wjvwevTfN7/uC/ltbvwRiTsAq7JQ2034vWxDgJvAE/4x98N7PDr/S3w+4VSWyFct1F1PgxsB7b5b+h5BfCeuxZv5MGreN1Xea0npa5leKNuXvDfX3mtDfguXvdi3H+v3Qo0AE8Bu4H/BOoLqLaCeK8Bl+J1+2xLybVrs712mvovIlIkAtXlIiIiY1Ogi4gUCQW6iEiRUKCLiBQJBbqISJFQoIuIFAkFuohIkVCgi/jMbI2/WFOlPxNzh5mdm++6RDKliUUiKczsHqASqAJanXOfz3NJIhlToIuk8Nds2QwMAG9zzo3kuSSRjKnLReRkDUAN3u4xlXmuRWRS1EIXSWFmG/B2JlqKt2DTHXkuSSRjBbceuki+mNnNQNw59x1/yeNfmdnvOed+mu/aRDKhFrqISJFQH7qISJFQoIuIFAkFuohIkVCgi4gUCQW6iEiRUKCLiBQJBbqISJH4/99QDqYCUT5ZAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEICAYAAABPgw/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAa3klEQVR4nO3de3xdZZ3v8c83O7emVyDRAVooSlWKqEAseAfhjOVii4MFyjDAiHBQ8TUqgszoQag6oziAgy+OCuowoFAuOhihyowMTI8I2DAg0GKhVLClXNJbGpp78jt/7F3YTXaa3XQnK3v1+3698ure63nyPL8nSb9ZWWvtvRQRmJlZ+atIugAzMysNB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaWEA91SS9J+kl6VlBmFsf9a0n+UelyzXSFfh25pIek54JMR8ZsSjzsT+BNQFRG9pRzbrJS8h25mlhIOdBuXJF0i6VlJbZJWSPpYXtu5kp7KaztM0k3AfsAvc4dZLpY0U1JIqpR0qqTmAXN8XlJT7vEJkh6VtEXSGkmX5XVdmvt3c27s90g6W9Jv88Z6r6Rlklpz/743r+1+SV+T9ECu5v+QVJ9rq5X0E0kbJG3Ofe4bS/4Ftd2CA93Gq2eBDwBTgcuBn0jaW9IC4DLgTGAKMA/YEBF/A/wZ+GhETIqIKwaM90vgrZJm5W07Hbg593hrbsxpwAnApySdlGv7YO7fabmxH8wfWNKewN3ANcBewFXA3ZL2GjDX3wJvAKqBL+a2n5Vb44zc554PdBT1FTIbwIFu41JE3B4R6yKiPyJuBZ4B5gCfBK6IiGWRtSoini9ivHbgF8BCgFywvw1oyrXfHxFP5OZ7HLgF+FCR5Z4APBMRN0VEb0TcAvwR+Ghen3+NiKcjogO4DXhXbnsP2SA/MCL6IuKRiNhS5Lxm23Gg27gk6UxJj+UOQ2wG3g7Uk92TfXaEw95MLtDJ7jHfmQt6JB0h6T5JLZJaye4p1xc57j7AwF8qzwP75j1/Ke9xOzAp9/gm4B5gsaR1kq6QVFX0iszyONBt3JG0P3A9cAGwV0RMA54EBKwB3jzEpw53ydZ/Ag2S3kU22G/Oa7uZ7N76jIiYCnw/N18x464D9h+wbT/ghWE+j4joiYjLI2I28F7gRLKHfsx2mgPdxqOJZEO0BUDS35LdQwf4IfBFSYcr68DcLwCAl4E3DTVoRPQAtwPfBvYkG/DbTAY2RkSnpDlk9+C3aQH6dzD2EuAtkk7fdgIWmA3cNdxCJR0t6ZDctfJbyB6C6R/u88wKcaDbuBMRK4ArgQfJhvQhwAO5ttuBb5Ddo24D7iQbzgD/BHwld5jmiwPHzbkZOBa4fcA15Z8GFklqAy4le5x7Wz3tuTkfyI195IB6N5Dds74Q2ABcDJwYEeuLWO5fAHeQDfOngP8mexjGbKf5hUVmZinhPXQzs5RwoJuZpYQD3cwsJRzoZmYpUZnUxPX19TFz5sykpjczK0uPPPLI+ohoKNSWWKDPnDmT5ubm4TuamdlrJA35Vhc+5GJmlhIOdDOzlHCgm5mlhAPdzCwlHOhmZinhQDczSwkHuplZSiR2HfquaG9vZ0H9uXR3dgNQN7mWX7T6HUfNbPc27B66pB9LekXSk0O0S9I1klZJelzSYaUvc3vzJ531WpgDtLd18r8qFoz2tGZm41oxh1xuAObuoP04YFbu4zzge7te1tBO2vOsIdvOfOtnR3NqM7NxbdhAj4ilwMYddJkP3Ji7A/tDwDRJe5eqwIG2bm4fsu3FZ14ass3MLO1KcVJ0X7I37t1mLdvf7fw1ks6T1CypuaWlpQRTm5nZNmN6lUtEXBcRjRHR2NBQ8M3ChlU3pXbItjfsXz/S0szMyl4pAv0FYEbe8+m5baPi4xfOG7rtCx8drWnNzMa9UgR6E3Bm7mqXI4HWiHixBOMWdPSp76O6tmrQ9qraKj5w8pEFPsPMbPdQzGWLtwAPAm+VtFbSOZLOl3R+rssSYDWwCrge+PSoVQtMf8s+nHHpAioyyhUIFZkKzr/qLOr32XM0pzYzG9eGfWFRRCwcpj2Az5SsomH09fZx/+IHqMhk6O/rhYBMZYZ7b1rKR//3XyJprEoxMxtXyu6l/w/f/T+8uPplert7X9vW09XDn574M3+4f3mClZmZJavsAv2Pv3+Gjlc7B23v7uxh5bJnE6jIzGx8KLtA/4sD3kjtxJpB26snVPFGX7ZoZruxsgv0o059L1U1VeQfKleFqK2r4b0nzUmuMDOzhJVdoNdNnsDVSxfxpnfOpKq6ksrqSt767gP5zm+/TnXN4MsZzcx2F2X59rn7z57B9//n22xuaaWiooIpe01OuiQzs8SVZaBvM61hatIlmJmNG2V3yMXMzApzoJuZpYQD3cwsJRzoZmYp4UA3M0sJB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaVEWQb6xpc2seiUKzl+wkJOqDudfzrjGlrXb0m6LDOzRJXde7l0d/Xw2SP/gQ3rNtLX2w/A0tt/x9PNq/jh8qvJZDIJV2hmloyy20P/7c8fpm3jq6+FOUBvTx8bXtzEsl89lmBlZmbJKrtAf375msK3oOvo4fkVaxOoyMxsfCi7QN/voOlMmFQ7aHv1hCpmvG2fBCoyMxsfyi7QP3DyEUycWkdF5vXSM1UZ9njDVI44/rAEKzMzS1bZBXp1bTXffegfOfLEw8lUZqisruT9H5vDdx74BplKnxA1s91X2V3lAlC/715c/u8XExEAKP+O0WZmu6myDPRtHORmZq8ru0MuZmZWmAPdzCwlHOhmZilRVKBLmitppaRVki4p0L6fpPskPSrpcUnHl75UMzPbkWEDXVIGuBY4DpgNLJQ0e0C3rwC3RcShwGnA/y11oWZmtmPF7KHPAVZFxOqI6AYWA/MH9AlgSu7xVGBd6Uo0M7NiFBPo+wJr8p6vzW3LdxlwhqS1wBLgs4UGknSepGZJzS0tLSMo18zMhlKqk6ILgRsiYjpwPHCTpEFjR8R1EdEYEY0NDQ0lmtrMzKC4QH8BmJH3fHpuW75zgNsAIuJBoBaoL0WBZmZWnGICfRkwS9IBkqrJnvRsGtDnz8AxAJIOIhvoPqZiZjaGhg30iOgFLgDuAZ4iezXLckmLJM3LdbsQOFfSH4BbgLNj2xutmJnZmCjqvVwiYgnZk5352y7Ne7wCeF9pSzMzs53hV4qamaWEA93MLCUc6GZmKeFANzNLCQe6mVlKONDNzFLCgW5mlhIOdDOzlHCgm5mlhAPdzCwlHOhmZinhQDczSwkHuplZSjjQzcxSwoFuZpYSDnQzs5RwoJuZpYQD3cwsJRzoZmYp4UA3M0sJB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaVEZdIFjFRfXx/PPLIaSRx42AFkMpmkSzIzS1RZBvof/ns5X1twJd1dPQDU1tVw2c8vYvZ73ppwZWZmySm7Qy5bNrTxlRO/Sev6NjraOulo62TTy61cMvfrbG3dmnR5ZmaJKbtAv2/xA0R//6Dt0R8sveOhBCoyMxsfyi7Qt6xvo6uze9D2nq5eWte3JVCRmdn4UFSgS5oraaWkVZIuGaLPKZJWSFou6ebSlvm6dx59MLV1NYO2V9VU8q6jDx6tac3Mxr1hA11SBrgWOA6YDSyUNHtAn1nA3wPvi4iDgc+NQq0AHPKBg3jnUQdTO/H1UK+dWMO75x7K2+bMGq1pzczGvWKucpkDrIqI1QCSFgPzgRV5fc4Fro2ITQAR8UqpC91GEpf/+8X85idLueeG+5DE3E98mA+f/v7RmtLMrCwUE+j7Amvynq8FjhjQ5y0Akh4AMsBlEfHrgQNJOg84D2C//fYbSb0AZCozfOTso/nI2UePeAwzs7Qp1UnRSmAWcBSwELhe0rSBnSLiuohojIjGhoaGEk1tZmZQXKC/AMzIez49ty3fWqApInoi4k/A02QD3szMxkgxgb4MmCXpAEnVwGlA04A+d5LdO0dSPdlDMKtLWKeZmQ1j2ECPiF7gAuAe4CngtohYLmmRpHm5bvcAGyStAO4DLoqIDaNVtJmZDaaISGTixsbGaG5uTmRuM7NyJemRiGgs1FZ2rxQ1M7PCHOhmZinhQDczSwkHuplZSjjQzcxSwoFuZpYSDnQzs5RwoJuZpYQD3cwsJRzoZmYp4UA3M0sJB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaWEA93MLCUc6GZmKeFANzNLCQe6mVlKONDNzFLCgW5mlhIOdDOzlHCgm5mlhAPdzCwlHOhmZinhQDczSwkHuplZSjjQzcxSoqhAlzRX0kpJqyRdsoN+J0sKSY2lK9HMzIoxbKBLygDXAscBs4GFkmYX6DcZ+Dvg4VIXaWZmwytmD30OsCoiVkdEN7AYmF+g39eAbwGdJazPzMyKVEyg7wusyXu+NrftNZIOA2ZExN07GkjSeZKaJTW3tLTsdLFmZja0XT4pKqkCuAq4cLi+EXFdRDRGRGNDQ8OuTm1mZnmKCfQXgBl5z6fntm0zGXg7cL+k54AjgSafGDUzG1vFBPoyYJakAyRVA6cBTdsaI6I1IuojYmZEzAQeAuZFRPOoVGxmZgUNG+gR0QtcANwDPAXcFhHLJS2SNG+0CzQzs+JUFtMpIpYASwZsu3SIvkftellmZraz/EpRM7OUcKCbmaWEA93MLCUc6GZmKeFANzNLCQe6mVlKONDNzFLCgW5mlhIOdDOzlHCgm5mlRFEv/R+Pon8LdC0FAmo+gCqmJV2SmVmiyjLQ+zt+Da0XgzIQAL3ElK9TUVfoRkpmZruHsjvkEn0t0HoR0AmxFdgKdMGWrxB96xKuzswsOWUX6HTeA6hAQ0Dnr8a6GjOzcaP8Ap0uoK/A9l6iv2OsizEzGzfKL9BrjqJw2dWo9sNjXIyZ2fhRdoGuyjdD3VnABLKHXpR9XLcAVc1OtjgzswSV5VUuFVO+SNQeQ3Q0AYEmnABVvie1me3eyjLQAVR9KKo+NOkyzMzGjbI75GJmZoU50M3MUsKBbmaWEg50M7OUcKCbmaWEA93MLCUc6GZmKeFANzNLCQe6mVlKONDNzFLCgW5mlhJFBbqkuZJWSlol6ZIC7V+QtELS45LulbR/6Us1M7MdGTbQJWWAa4HjgNnAQkkD36f2UaAxIt4B3AFcUepCzcxsx4rZQ58DrIqI1RHRDSwGtrsbc0TcFxHtuacPAdNLW6aZmQ2nmEDfF1iT93xtbttQzgEK3txT0nmSmiU1t7S0FF+lmZkNq6QnRSWdATQC3y7UHhHXRURjRDQ2NDSUcmozs91eMTe4eAGYkfd8em7bdiQdC3wZ+FBEdJWmPDMzK1Yxe+jLgFmSDpBUDZwGNOV3kHQo8ANgXkS8UvoyzcxsOMMGekT0AhcA9wBPAbdFxHJJiyTNy3X7NjAJuF3SY5KahhjOzMxGSVH3FI2IJcCSAdsuzXt8bInrMjOzneRXipqZpYQD3cwsJRzoZmYp4UA3M0sJB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaVEUa8UHY8i+qDnEejfCtWHo4opSZdkZpaosgz06HmK2HQORAcgiB5i8t9TMfH0pEszM0tM2R1yieglNn0C+tdDbIV4FeiCtm8SPU8kXZ6ZWWLKLtDp/j1EZ6EGov3WMS/HzGy8KL9AjzZABRr6oX/zWFdjZjZulF+gV78bortAwwRU+5ExL8fMbLwou0BXxZ4w6XNALa/vqU+AqoOgdm6ClZmZJassr3KpmPRJovpQon0x9LeiCcdB7QlIVUmXZmaWmLIMdABVH46qD0+6DDOzcaPsDrmYmVlhDnQzs5RwoJuZpYQD3cwsJRzoZmYpUbZXuUR0Q+cSovshqNgH1S1Amb2TLsvMLDFlGejR/yqx4RToXwfRDlQT7T+CadehmiOSLs/MLBFlecgltv4r9K3JhTlAN0QH0XohEf2J1mZmlpSyDHQ67wa6Bm/vb4O+58a6GjOzcaE8A101QzR0EX1rx7QUM7PxojwDfcLpwIQCDf2w6QL6O+56bUv0t9G/9Qb6N19E/6s/JPo3jWpp0fMU0XkP0fv8qM5TbqL3eaL790R/a9Kl7NbWPfsSy3+3ko5XO5IuZUx1tnex4sGVvLDqxZKP/dJzr7DiwZW0t23/Ne3Y2smKB1dy70/uZ/4eZ7Jg70+ycePGks+fTxExfCdpLvAvQAb4YUR8c0B7DXAjcDiwATg1Ip7b0ZiNjY3R3Nw8oqIj+ojWi6FzCdBXoOCJ6A0PQ38Lsf7k3K3qOoBaUDXa61ZU+eYRzT1kTf1txKZPQs8fQRmIHqj5EJp29W79pmHRv5nY9CnoWQ6qguiCieegSZ9DKvS+9jYaNre08tWTruDZx54jU5Whr6ePsxadyoIL5yVd2qj7xbW/4vov/ZRMZQV9PX0c8I79WXTnxezxxmm7NG7bple5/OR/5qmHn6GqupLe7l7OuPTjnPalj/Hzf7mLH/3DLXR3DH6r7yn1k/nZKz8e8bySHomIxkJtw+6hS8oA1wLHAbOBhZJmD+h2DrApIg4Erga+NeJqiyBlqJh2JQwZyoKeFcSWf4TYTDbMAToh2ojWS0teU2z5P9DzZHaubbfF61pKbP1ByecqJ7H5Quh5nG1fe+iG9hty50FsrFz+8St5uvlZujq6ad/SQVdHNzd+9TZ+/6tHky5tVD36X09w/Zd+Sld712vrfuaR1Vx60hW7PPY3Trua5b9bSXdHN1tb2+nq6OanX/8ZN1x6Kz/+8uKCYQ6wZX0bTd/79S7PX0gxh1zmAKsiYnVEdAOLgfkD+swH/i33+A7gGI3F7ldFQ+Ht0QcVk6Dr/wEDr3oJ6HmEiAJ79iOUvSb+P4GeAS2d0H5LyeYpN9G/EbofZtDXJTqIrSPfQ7Gd88qfW3h62Sp6e7b/me9s7+L2K5sSqmps3HHVXXS1b38BRV9vH396/PldOvyy6eXNPL70KXq7e7fb3rm1izu/u2TQnAN99zM/GvHcO1JMoO8LrMl7vja3rWCfiOgFWoG9Bg4k6TxJzZKaW1paRlZx/nh1f8PgY+kVkNkHMm/O/olfUIbCt7Eboehl8C+ObW2717HK7fS3kf1aF2ob3XMZ9rrW9W1UVhd+ycmml9J928aNLxX+OctUZWht2TLicbdsaKOyqvDP9nBhPprG9KRoRFwXEY0R0djQMMTe9c6oORomngVUgyaBJkLF3miPH2SPz074WLZtO1VQOxepdEtXRR1UHligpQKq31+yecpOZjqotkBDJdR8cMzL2V3tf/AMon/wubLK6krmHH9YAhWNnSOOP4yqmsG/zPp6+zngHfuPeNx9Z+1NReXgDMlUZdj/4Blkhgj7babUTx7x3DtSTKq9AMzIez49t61gH0mVwFSyJ0dHlSQqJn8BNdyHpn4L7XE9argXVe6XbZ98IVS9AzQBVJf9qJyFpny19LVM/UZ2fLb9VVADmoomX1zyucqFlIEpl7P97QKrQVPQpE8nWNnupbqmik9dfTY1da9f7ltVU8XkPSdxykXpPin6V393AlPqp2wX6jV1NZz7rb9mwsRCOxvFqayq5IJrPkFN3es7jFXVlUyaWsdFN3yGKXvtOLBvXvO9Ec+9I8Ne5ZIL6KeBY8gG9zLg9IhYntfnM8AhEXG+pNOAv4qIU3Y07q5c5bKzoucJ6HkGKmdC1aGjdnVF9K0j2n8Kvaug6l2o7tTsPVB3c9HzRPaYed9aqH4PqjsLZQYdkbNR9vjSFdxx1S9Zv3YjjR95Jyd//kSm1k9JuqxRt2VDGz//zl08vORR9tpnD07+/Ikc+uFDSjL2kw/8kdv/uYmXn2/hsGPfwYILP8oeb5zG5pZWfnb1Xfz6xnvZvK5tu8+58fnvsfeM+hHPuaOrXIq9bPF44DtkD4j+OCK+IWkR0BwRTZJqgZuAQ4GNwGkRsXpHY45loJuZpcWOAr2oN+eKiCXAkgHbLs173Aks2JUizcxs15TnK0XNzGwQB7qZWUo40M3MUsKBbmaWEg50M7OUcKCbmaWEA93MLCWKemHRqEwstQCluAtEPbC+BOOUC683vXantYLXO1L7R0TBN8NKLNBLRVLzUK+aSiOvN712p7WC1zsafMjFzCwlHOhmZimRhkC/LukCxpjXm16701rB6y25sj+GbmZmWWnYQzczMxzoZmapUTaBLmmupJWSVkm6pEB7jaRbc+0PS5o59lWWRhFr/YKkFZIel3SvpJHfHHEcGG69ef1OlhSSyvpSt2LWK+mU3Pd4uaSbx7rGUiri53k/SfdJejT3M318EnWWgqQfS3pF0pNDtEvSNbmvxeOSSntT14gY9x9k75T0LPAmsnd9/gMwe0CfTwPfzz0+Dbg16bpHca1HA3W5x58q17UWu95cv8nAUuAhoDHpukf5+zsLeBTYI/f8DUnXPcrrvQ74VO7xbOC5pOvehfV+EDgMeHKI9uOBX5G9ye6RwMOlnL9c9tDnAKsiYnVEdAOLgfkD+swH/i33+A7gGI3WzUNH17BrjYj7IqI99/QhsjfuLlfFfG8BvgZ8C+gcy+JGQTHrPRe4NiI2AUTEK2NcYykVs94Att3cdCqwbgzrK6mIWEr2NpxDmQ/cGFkPAdMk7V2q+csl0PcF1uQ9X5vbVrBPRPQCrUA53om4mLXmO4fsb/xyNex6c3+WzoiIu8eysFFSzPf3LcBbJD0g6SFJc8esutIrZr2XAWdIWkv2VpefHZvSErGz/793SlH3FLXxSdIZQCPwoaRrGS2SKoCrgLMTLmUsVZI97HIU2b++lko6JCI2J1rV6FkI3BARV0p6D3CTpLdHRH/ShZWbctlDfwGYkfd8em5bwT6SKsn+6bZhTKorrWLWiqRjgS8D8yKia4xqGw3DrXcy8HbgfknPkT3u2FTGJ0aL+f6uBZoioici/gQ8TTbgy1Ex6z0HuA0gIh4Easm+kVUaFfX/e6TKJdCXAbMkHSCpmuxJz6YBfZqAs3KPPw78V+TOQpSZYdcq6VDgB2TDvJyPr8Iw642I1oioj4iZETGT7DmDeRHRnEy5u6yYn+U7ye6dI6me7CGY1WNZZAkVs94/A8cASDqIbKC3jGmVY6cJODN3tcuRQGtEvFiy0ZM+K7wTZ4+PJ7un8izw5dy2RWT/c0P2h+B2YBXwe+BNSdc8imv9DfAy8FjuoynpmkdzvQP63k8ZX+VS5PdXZA8zrQCeAE5LuuZRXu9s4AGyV8A8Bvxl0jXvwlpvAV4Eesj+pXUOcD5wft739trc1+KJUv8s+6X/ZmYpUS6HXMzMbBgOdDOzlHCgm5mlhAPdzCwlHOhmZinhQDczSwkHuplZSvx/hWYv5Mpc3OQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/explore/miniconda3/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/loss.py:446: UserWarning: Using a target size (torch.Size([])) that is different to the input size (torch.Size([1])). This will likely lead to incorrect results due to broadcasting. Please ensure they have the same size.\n",
      "  return F.mse_loss(input, target, reduction=self.reduction)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEGCAYAAAB1iW6ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3de5zcdX3v8ddnbjuztyTkQsNurhCVBRHCkqKCaKGKqASoYKgWPNhCT6UPD/YcH+nxlArl8aCKrT0KRw+tWooiXk6VVFFqFcpNLgk3CZFkuWaTSDbJJnuZy87Mfs8fv98kk81udnZmkpnfzPv5eOwjM7/fb3c/O8y++e7n9/39vuacQ0REgi9U6wJERKQ6FOgiIg1CgS4i0iAU6CIiDUKBLiLSICK1+sbz5s1zS5curdW3FxEJpA0bNuxyzs2fbF/NAn3p0qWsX7++Vt9eRCSQzOy1qfap5SIi0iAU6CIiDUKBLiLSIGrWQxeR4Mtms/T395NOp2tdSsOJx+N0d3cTjUZL/hwFuoiUrb+/n46ODpYuXYqZ1bqchuGcY/fu3fT397Ns2bKSP2/alouZfcPMdprZ81PsNzP7spn1mdlzZrZyBnWLSICl02nmzp2rMK8yM2Pu3Lkz/sunlB76PwPnH2b/+4EV/sfVwFdnVIGIBJrC/Mgo53WdtuXinHvQzJYe5pDVwL847z68j5nZbDNb6JzbMeNqRMrxwAPwy1/WuoryhEJwxRWwfHmtK5EGUI0eehewteh5v7/tkEA3s6vxRvEsXry4Ct9aBPjMZ+DJJyGII0XnYHQUbrml1pXIJB544AG++MUv8uMf/3jaY7/yla9w2223EQ6H+cAHPsAXvvAFAG6++Wa+/vWvEw6H+fKXv8z73vc+AH72s5/xqU99inw+zx//8R+zdu3aius9qidFnXO3A7cD9Pb2amUNqY7hYbjsMvjud2tdycwdeyyMjNS6CqnQ/fffzz333MOzzz5LS0sLO3fuBOCFF17g7rvvZuPGjWzfvp3zzjuPzZs3A/DJT36Sn//853R3d3PGGWdw4YUX0tPTU1Ed1ZiHvg1YVPS8298mcnQkk9DaWusqytPa6tUvZfvWt77FqlWrOPXUU7nmmmvI5/MAtLe3c91113HSSSdx7rnnMjAwAMAzzzzDmWeeySmnnMLFF1/M4OAgAH19fZx33nm87W1vY+XKlbz00ksAjIyM8OEPf5i3vOUtfPSjH2WyVd6++tWvsnbtWlpaWgBYsGABAPfccw9r1qyhpaWFZcuWccIJJ/DEE0/wxBNPcMIJJ7B8+XJisRhr1qzhnnvuqfi1qMYIfR1wrZndDfwusE/9czmqFOh14YZ/28gL24eq+jV7juvkrz900pT7N23axHe/+10eeeQRotEof/Znf8a3v/1trrjiCkZHR+nt7eVLX/oSN954IzfccAO33norV1xxBV/5ylc455xzuP7667nhhhv4h3/4Bz760Y+ydu1aLr74YtLpNOPj42zdupWnn36ajRs3ctxxx/HOd76TRx55hLPOOuugOjZv3sxDDz3EZz/7WeLxOF/84hc544wz2LZtG2eeeeb+47q7u9m2zRvvLlq06KDtjz/+eMWv17SBbmbfAd4NzDOzfuCvgSiAc+5rwL3ABUAfkAT+S8VVicyEAr1p/eIXv2DDhg2cccYZAKRSqf2j41AoxEc+8hEAPvaxj3HJJZewb98+9u7dyznnnAPAlVdeyaWXXsrw8DDbtm3j4osvBryLegpWrVpFd3c3AKeeeiqvvvrqIYGey+XYs2cPjz32GE8++SSXXXYZL7/88pH94SdRyiyXy6fZ74BPVq0ikZlwruJA7x9MsjeZJTmWZ3QsR2osz2gmRyqbZzSTZyw3TltLmI54hPaWKO3xCO0tEf95hGPaYsSj4fK+eQMF+uFG0keKc44rr7ySm2++edpjy51eWWijAITDYXK53CHHdHd3c8kll2BmrFq1ilAoxK5du+jq6mLr1gNzRvr7++nq6gKYcnsldKWoBFvhwosZBvobQ2nWPbOdf316G5t2VNYmaIuFubR3EVe8fQnL57fP7JNbW2HPnoq+fzM799xzWb16Nddddx0LFixgz549DA8Ps2TJEsbHx/nBD37AmjVruOuuuzjrrLOYNWsWc+bM4aGHHuLss8/mzjvv5JxzzqGjo4Pu7m5+9KMfcdFFF5HJZPb34ktx0UUXcf/99/Oe97yHzZs3MzY2xrx587jwwgv5wz/8Qz796U+zfft2tmzZwqpVq3DOsWXLFl555RW6urq4++67ueuuuyp+PRToEmyF0W0JgT6ayXHfxt/yw6e38UjfLsYdvG3RbP7qgz0smpOgrSVCIhamLRahNRb2PyJEw0Yym2c4nWMknWMkk/UeZ7znj7+yh28//hr//OirvPvN8/n4O5byrhXzCYVKGBG2tkJ/f4UvQvPq6enhpptu4r3vfS/j4+NEo1Fuu+02lixZQltbG0888QQ33XQTCxYs4Lv+LKg77riDP/3TPyWZTLJ8+XK++c1vAnDnnXdyzTXXcP311xONRvn+979fch1XXXUVV111FSeffDKxWIw77rgDM+Okk07isssuo6enh0gksn9aI8Ctt97K+973PvL5PFdddRUnnVT5Xzg22Rnbo6G3t9dpgQup2NatsHgx/NM/wSc+MekhT70+yL88+ir3bXyDVDZP95wEF5/WxUWndXH8TEfUU9g5nOY7j2/lW4+/xsBwhuXz2/j4O5Zyycpu2lsOM276oz+CRx8Ff0ZF0GzatIkTTzyx1mVMqr29nZGATwmd7PU1sw3Oud7JjtcIXYJtmhH6cDrL5bc/RkskxEWndXHJyi56l8yp+uXqCzrifOq8FfzXdx/PT5/fwTcfeZXr79nILT97kes/1MOlvYsm/8QG6qFL7SnQJdhGR71/29om3f3oS7vJ5Ma546pVnLl87hEvJxYJsfrULlaf2sXTrw/yP3/4PLfe3zd1oLe1HfgZpKqCPjovhxa4kGCbZoT+4OYB2mJhVi6ecxSL8py2eA4Xn3Ycr+1OsmskM/lBhRF6jVqf1VCrtm2jK+d1VaBLsB0m0J1zPLhlgLcfP49YpDZv9cL/SJ56bXDyA1pbIZ+HbPYoVlU98Xic3bt3K9SrrHA/9OL58KVQy0WC7TCB/uruJFv3pLj67NrdyfDkrllEw8aG1wd570m/c+gBhbqTSYjFjm5xVdDd3U1/f//+y+qlegorFs2EAl2C7TCB/uBmL2Te9ab5R7Oig8SjYU7umnX4ETp4P8fs2UevsCqJRqMzWlFHjiy1XCTYpgn0JXNbWTJ38hOmR8vpi+fwXP8+xnLjh+4sDnSRCinQJdimCPRMLs+jL+3mnBqOzgtWLplDJjfOC5NdkapAlypSoEuwTRHoG14dJJXN864VtQ/005d4J0Y3TNZ2UaBLFSnQJdiSSW+loqIbKAH855YBomHj7ccf+bnn0zm2M07X7ARPva5AlyNLgS7BNjrqheKEKz8f3LyL05fMoe1wl90fRSuXzJn8xGgh0HVxkVSBAl2CLZk85CrRncNpNu0YqunslolOXzybHfvSbN+bOnhHoXaN0KUKFOgSbJPcC/2hzbsA6qJ/XnD6kmMADm27qOUiVaRAl2CbJND/c/MA89pb6FnYWaOiDvWWhR0kouFDT4wq0KWKFOgSbBMCfXzc8XDfLt61Yl5p9yM/SqLhEKd0T3KBkQJdqkiBLsE2IdCf376PPaNjddU/Lzh9yRw2bh8inS1aCSeR8P5VoEsVKNAl2CYEeuFy/7NWzKtVRVM6fckccuOO5/r3HdgYCkE8rkCXqlCgS7AdEui7OLmrk3ntLYf5pNo4bfEUFxhpkQupEgW6BFtRoA+lszz1+mBdzW4pdkxbjOXz2iaf6aJAlypQoEuwFQX6o327yY27uuyfFxQuMDro/uEKdKkSBboEW+FKUeDBLQO0t0RqsjpRqVYunsPu0TFe210U4K2tulJUqkKBLsHl3P4rRZ1zPLh5gLcfP7dmqxOVonCjroPaLm1tGqFLVdTvO19kOtmst3xbayuv7BqlfzBV1+0WgBUL2uloiRx8YlQtF6kSBboEV9GtcwvTFc+p0xOiBaGQcdqSOQp0OSIU6BJcxYG+ZRdL57ayeO6hKxfVm5WLZ7P5jWGG0/7C0Ap0qRIFugSXH4LZlji/eml33bdbCk5fModxB89u9S8wUqBLlSjQJbj8EOwbGSeVzdfFcnOlOHXRbMyKLjBSoEuVKNAluPwQfG73GNGwceby2q9OVIqOeJQ3H9txYKaLAl2qpKRAN7PzzexFM+szs7WT7F9sZveb2dNm9pyZXVD9UkUm8EPw14NZTltUP6sTlWLlkjk89fog4+POC/RMxpuxI1KBaQPdzMLAbcD7gR7gcjPrmXDY/wK+55w7DVgD/J9qFypyCP9inIHxCMfOite4mJk5ffEchtM5+gZGdAtdqZpSRuirgD7n3MvOuTHgbmD1hGMcUFhNYBawvXolikzBD8BdLsKsRHBG53DgAqMNrw1qGTqpmlICvQvYWvS8399W7HPAx8ysH7gX+PPJvpCZXW1m681s/cDAQBnlihTxA3AgH6YzHq1xMTOzZG4rx7TFvAUvNEKXKqnWSdHLgX92znUDFwB3mtkhX9s5d7tzrtc51zt/fjBmJEgd8wNwJByjMxGsQDczVi6ew4bXFehSPaUE+jZgUdHzbn9bsU8A3wNwzv0KiAP1t8KANBY/AFORlsCN0MFru7w8MMpwyK9dgS4VKiXQnwRWmNkyM4vhnfRcN+GY14FzAczsRLxAV09Fjiw/ANPRGJ0B66GDd8UowJYRf3aLAl0qNG2gO+dywLXAfcAmvNksG83sRjO70D/sL4A/MbNnge8AH3cH3fBZ5AhIJhlvieMsFMgR+indswmHjN/sU6BLdZQ0rHHO3Yt3srN42/VFj18A3lnd0kSmkUySj3uLLAethw6QiIWZ0xplYDzsbVCgS4V0pagEVzJJNu7NP58VwEAH739Eg84fVynQpUIKdAmu0VGyMS/QO+PB66EDdMaj7Br3a9eqRVIhBboEVzLJmB/oHQHsoYM3Qt+tEbpUiQJdgiuZJN0SJxEN1/Wyc4fTGY+wK68eulRHMH8LRACSSW8OegCnLBbMSkQZzDqIRhXoUjEFugRXMkkyoBcVFXQmouxLZXG6ha5UgQJdgiuZJBkJ3mX/xTrjUbJ5p3uiS1Uo0CW4kkmGQ7HATlmEA9Mt8/GEAl0qpkCX4PIDPahTFoH9/X8FulSDAl2CK5lkKBQNfMsFvIWuFehSKQW6BFM+D5kM+4gG/qQo+IGuC4ukQgp0CSZ/NJtsgGmLAJmoRuhSOQW6BFPhXuixeLBH6H7/PxVToEvlFOgSTMWLWwS4h164ZUEq2qJAl4op0CWYCoEebQn0tMVYJEQiGiYZiSnQpWIKdAmmokAPcssFvD76aEiBLpVToEswFZafC/hJUfDmog+H/ZaLFvqSCijQJZgaaITeGY8eWCg6na5tMRJoCnQJpqJA7wjwlaLgtVz2mR/oartIBRToEkz+RTiWSBAJB/tt3JmIshetWiSVC/ZvgjQvfyQb6WivcSGV64xHGEQjdKmcAl2CyQ++aCMEuhaKlipRoEsw+cEXn9VR40IqNysRZTTqrY2qQJdKKNAlmJJJsuEIbe3xWldSsc54lFSkxXuiQJcKKNAlmJJJ0g0wZRG8eejpqAJdKqdAl2Dav0B0IwR61LuXCyjQpSIKdAkkN1pYIDrYc9BBLRepHgW6BFJuZMS7SrQBRuizNEKXKlGgSyDlh0e8+7g0Qg89Hj3QQ9eFRVIBBboEUn402TAj9I54hLFIFGemEbpUpKRAN7PzzexFM+szs7VTHHOZmb1gZhvN7K7qlilyMDc66gd68HvooZDRHo9qoWip2LS/DWYWBm4Dfh/oB540s3XOuReKjlkB/CXwTufcoJktOFIFiwDeLJdYJ8c2QMsFvLZLJhYnpkCXCpQyQl8F9DnnXnbOjQF3A6snHPMnwG3OuUEA59zO6pYpcjBLpUhF44FerahYZyJKOpbQCF0qUkqgdwFbi573+9uKvQl4k5k9YmaPmdn5k30hM7vazNab2fqBgYHyKhYBQqlUQ9wLvWBW4eIiBbpUoFonRSPACuDdwOXAP5rZ7IkHOedud871Oud658+fX6VvLc0okvYCvb0B5qFDYS66lqGTypQS6NuARUXPu/1txfqBdc65rHPuFWAzXsCLVJ9zRDMpxuMJwiGrdTVV0ZmIMhrWCF0qU0qgPwmsMLNlZhYD1gDrJhzzI7zROWY2D68F83IV6xQ5wF+mzbUmalxI9cxKRBkJa4QulZk20J1zOeBa4D5gE/A959xGM7vRzC70D7sP2G1mLwD3A//DObf7SBUtTc4PPWttq3Eh1dMZ9wLdKdClAiU1IJ1z9wL3Tth2fdFjB3za/xA5sgrLz7W11riQ6ulMREhFWxgfGSFc62IksHSlqASPP4oNtzXWCD0dUQ9dKqNAl+ApBHp74wT6/ht0JVO1LkUCTIEuwVNYILoz+OuJFnQmoiSjcUKpJDhX63IkoBToEjj5Ea+HHmuABaILCj10y+chm611ORJQCnQJnPS+YQBaGmCB6IL9PXRQH13KpkCXwEnt9QI90UCBrkUupBoU6BI4maERABKzGqfl0hoLk4kp0KUyCnQJnMyQN0JvndNZ40qqx8ywVn9evQJdyqRAl8DJDXsj9PbZjRPoUDQNU8vQSZkU6BI4ueFRxjE6ZjfOPHQoCnSN0KVMCnQJnPyIv/xca6zWpVRVpN0/J6BAlzIp0CVwxv31RNtjjXEv9IJop0boUhkFugSOSybJxOKEGuRe6AUtnf40TAW6lEmBLsGTTDIWi9e6iqrbf6GUAl3KpECXwAmlUmRbGmdxi4LChVJZfxaPyEwp0CVwQukk+XjjjdBbZ3uBXrhwSmSmFOgSOJF0mnyi8Ubona0x0pGYRuhSNgW6BE40ncLFG2e1ooLORJRUpIWsRuhSJgW6BE5sLI1rbcBAj3s36MqPKNClPAp0CZRcfpx4Nt1Q64kWFO64mB/RLBcpjwJdAmUonSOezRBqwEDvTERIR1pwmrYoZVKgS6AMJcdIZDMNtZ5oQaHlonnoUi4FugTK8HCSiBsn0tE4i1sUxKNh0rE4pkCXMinQJVBGBr17oUcbaD3RYrl4glA6VesyJKAU6BIoyb1DAMQ6G6/lApCPxwkr0KVMCnQJlMJ6ovHOxmu5AIwnWolmFOhSHgW6BEp6nzdCT8xuzEAn0UpMI3QpkwJdAiWzzxuht3Q2Zg+d1lZiY5laVyEBpUCXQBkb9tbbtLbG7KGH2lqJ5bOQy9W6FAkgBboEyv4bVzXgpf8AIf9/VLq4SMqhQJdAyY94I/RGDfRIhxfoSf/kr8hMlBToZna+mb1oZn1mtvYwx/2BmTkz661eiSIH5Ef9kWuDtlyi/uyd0cGhGlciQTRtoJtZGLgNeD/QA1xuZj2THNcBfAp4vNpFihS4Bh+hFy6YGtUIXcpQygh9FdDnnHvZOTcG3A2snuS4vwE+D6SrWJ/IwVL+CL1BAz3uz95JaoQuZSgl0LuArUXP+/1t+5nZSmCRc+4nh/tCZna1ma03s/UDAwMzLlZk/42rGnDFIoC4P78+tU8jdJm5ik+KmlkI+HvgL6Y71jl3u3Ou1znXO3/+/Eq/tTSZTC5PNJ0iF2uBUGOezy8sFJ1RoEsZSvmt2AYsKnre7W8r6ABOBh4ws1eBM4F1OjEq1TaczpHIZcjHG3N0DtA2uxOAzJACXWaulEB/ElhhZsvMLAasAdYVdjrn9jnn5jnnljrnlgKPARc659YfkYqlaQ2lsiSymYZcILqgbY43Qs/6F1CJzMS0ge6cywHXAvcBm4DvOec2mtmNZnbhkS5QpGAonSORzeASjXlCFCDiz3LJjSjQZeYipRzknLsXuHfCtuunOPbdlZclcqihVJZ4LoM16AwXYP/snXEFupShMc8sSUMaSnstl0ZcIHo/P9DzCnQpgwJdAmMo5bVcQu0NeqdFgGiUXDiidUWlLAp0CYx9qSyJXIZIAy4QXWwsFlegS1kU6BIYQ2kv0EMNHujZeFzrikpZFOgSGEOpLK25scY+KQrkWxKEUhqhy8wp0CUwhtI5WrPphr2PS0E+kSCaTjE+7mpdigSMAl0CYyiVJZ7NNHygu0QriWyG4YxWLZKZUaBLYIyMpojmso0f6K2txHMZhlLZWpciAaNAl8DIDDX2vdALrNUboe9ToMsMKdAlMAoLRDd6oIfa2khkMwylFegyMwp0CYzciL9AdIMuP1cQ7mjzWy7qocvMKNAlENLZPOGUPze7wUfokfZ2b4SulovMUEk35xKptcJFRUDDB3qss51wTi0XmTkFugRC4T4uQOMHekcbls0wlByrdSkSMGq5SCB4d1r01x9v8EA3/xxBUsvQyQwp0CUQCqsVAQ0f6IWfL71vpMaFSNAo0CUQCndaBJom0DNDCnSZGQW6BEJh+TmgaQJ9bFiBLjOjQJdA2H8fF2iaQM8q0GWGFOgSCEPpLO3j/qyPJgn0/LCWoZOZUaBLIAylcsxyWYhGvY9G5s9yyY8q0GVmFOgSCEPprBfojT46h/0/YyiVIpsfr3ExEiQKdAmEoVSWjvxYUwW6bqErM6VAl0AYSudob7JAbx1LM5TWDbqkdAp0CYShVJa2Jgv0hEboMkMKdAkEb4Hoxl9+DjgQ6FrkQmZIgS51zznHULo51hMFoKUFZ0Zci1zIDCnQpe6ls+Nk8474WLo5At0M19rqt1zUQ5fSKdCl7hVGqbFmCXS0rqiUR4Euda9wYjDaRIFOayutuTG1XGRGSgp0MzvfzF40sz4zWzvJ/k+b2Qtm9pyZ/cLMllS/VGlWhVCLpNMNv55ogbW10enGNMtFZmTaQDezMHAb8H6gB7jczHomHPY00OucOwX4AfCFahcqzavQdginU001Qm/Pj2keusxIKSP0VUCfc+5l59wYcDewuvgA59z9zrmk//QxoLu6ZUozG0rlMDdOOJVsqkBvy6mHLjNTSqB3AVuLnvf726byCeCnk+0ws6vNbL2ZrR8YGCi9SmlqQ+ksLbkmudNiQaGHrkCXGajqSVEz+xjQC9wy2X7n3O3OuV7nXO/8+fOr+a2lgTXV8nMFhWmLOikqMxAp4ZhtwKKi593+toOY2XnAZ4FznHOZ6pQn4t3HZTZ+sDVRoMfH0hqhy4yUMkJ/ElhhZsvMLAasAdYVH2BmpwH/F7jQObez+mVKM3tt9yjdLf6TJgr0WDbNUCqHc67W1UhATBvozrkccC1wH7AJ+J5zbqOZ3WhmF/qH3QK0A983s2fMbN0UX05kRnL5cR59aTdnLIh7G5op0DNpxvLjZHK6J7qUppSWC865e4F7J2y7vujxeVWuSwSAX2/bx3A6x+lL/SF6EwV6JJMG57z1VKPhWlckAaArRaWuPbxlF2bw1mP8Zeea5MIi2toI5fNEx3OauiglU6BLXXuobxcnHddJ53jznRQF7xa6mukipVKgS90azeR4+vVBzjphPhQWTG66QE/rjotSMgW61K0nXtlDNu84e8U8SPoXIjddoGd4aWCkxsVIUCjQpW49tGUXLZEQpy+Z07SBfuKsCD99/rc1LkaCQoEudevhvgFWLTvGm+HRpIF+7uJ2Nrw2yI59qRoXJEGgQJe6tHMozeY3RjjrhHnehmQSzKCl5fCf2Cj8QH/Hcd6/9/5ao3SZngJd6tLDfbsAeGdxoLe2eqHeDPxAPy7mOHFhJ/f+ekeNC5IgUKBLXXp4yy6OaYvRs7DT25BsolvnwoGfNZnkg6csZMNrg2zfq7aLHJ4CXeqOc46H+3bxjuPnEgr5I/ImDvQL3roQQCdHZVoKdKk7W3aOsHM4401XLEgmm+cqUTjwsyaTLJvXRs/CTn7y3Pba1iR1T4EudeehLV7//KwVRffMHx1tzhG6f0HVB05ZyFOv71XbRQ5LgS515+EtAyyb10bX7MSBjc3Wckn4P7s/XbPQdtHJUTkcBbrUlbHcOI+/sufAdMWCZgv0UAji8f2BvmxeGycd18lPFOhyGAp0qStPvz5IcizPWSuaPNDB+3kLF1ThjdKffn0v29R2kSko0KWuPNy3i5DBmcvnHrxDgc4HCrNdNEqXKSjQpa48tGUXb1s0m1mJ6ME7FOgsVdtFpqFAl7qxL5Xluf69nD2xfw4KdN8HTlHbRaamQJe68auXdjPuii73L3BOge5T20UOR4EudePhvgFaY2FOWzzn4B3ZLOTzCnRgydw2Tu7q5MfPKdDlUAp0qRuP9O3mzOVziUUmvC0LodZMV4qC9/NOCHTwZrs8s3Uv/YOH7pPmpkCXutA/mOSVXaOHtlug+ZafK2htPfCzFznQdtG9XeRgCnSpCw/7l/ufPXH+OTTf4hYFk7RcwGu7vLVrFj9WH10mUKBLXXiobxcLOlpYsaD90J0K9ENc8NaFPLt1L1v3qO0iByjQpebGxx2P9u3irBPmYZMtYKFAP8T+tsvzGqXLAQp0qblfvbybwWT20Mv9C5o50MfGIJc7ZNfiua28tWsWP1EfXYpEal2ANKf+wSQ/fm4H//bsdjZuH6KjJcLZxbfLLdbMgQ6QSkFHxyG7P3DKQv72p7/h0q89yrvfvID3vHkBJy7smPyvHGkKCnQ5anYOpfnJr70Qf+r1vQCcumg2f/XBHj54ykLmd0yxAHSzB3oyOWmgf/wdS0mN5fnFb97glvte5Jb7XuR3OuO85y3zefebF/DOE+bR3qJf8WYSuP/av92XZttenQgqh3PT7J/kOOc/cf525xx55xh3MO4c4+Pe4/y4I53NM5LJMZrJMeJ/jGZyjGbyvDGUZsPrgzgHJy7s5DPnv5kPvvU4Fs8tIaQV6JPujkfDXPf7b+K6338TO4fSPLB5gPt/s5N/e3YH33liK9Gw0bOwk/Z4hNZYhNZY2P/wHidiYWLhECEzQgahkGGFx/6/RtFof5KH0/01oL8VJnfq4tkcP3+SCQAVClyg3/PMNm7+6W9qXYZMwwzaYhHaWyK0tYTpTET5899bwYdOWciKY00DX10AAAdPSURBVA8dbR5WM19YBFMGerEFnXEu613EZb2LyObHWf/qIPe/uJNNO4ZIjuXZM5oiOZYjOZYnmcmRzOan/R+8HDk3XXSyAh286VonFlaClxmbrr1aGJHZFKOxwkgudNBIzgiFIBEN+wEeIRENH1jguVLNfGERTHpx0eFEwyHefvxc3n783CmPcc6Rzo4zlh+Hwl9b/l9ezv83X5T47qDHpdWh/2FMbU5bdPqDylBSoJvZ+cD/BsLAPznn/nbC/hbgX4DTgd3AR5xzr1a3VM+iY1pZdEyT/WI3u8IINZE4/HGNZpqWSyXMjEQsTIJw1b+21M600xbNLAzcBrwf6AEuN7OeCYd9Ahh0zp0AfAn4fLULlSaWTHrLsYWabJbtEQx0aUyljNBXAX3OuZcBzOxuYDXwQtExq4HP+Y9/ANxqZubcEfij6xvfgL/7u6p/WaljO3Y03+gcDgT6NddAp9qMDeX66+EjH6n6ly0l0LuArUXP+4HfneoY51zOzPYBc4FdxQeZ2dXA1QCLFy8ur+K5c6Fn4h8I0tB6euDMM2tdxdH3lrd4Yb57d60rkWqbM2f6Y8pwVE+KOuduB24H6O3tLW/0vnq19yHS6KJR+NrXal2FBEgpTcltwKKi593+tkmPMbMIMAvv5KiIiBwlpQT6k8AKM1tmZjFgDbBuwjHrgCv9xx8GfnlE+uciIjKlaVsufk/8WuA+vGmL33DObTSzG4H1zrl1wNeBO82sD9iDF/oiInIUldRDd87dC9w7Ydv1RY/TwKXVLU1ERGaiySb2iog0LgW6iEiDUKCLiDQIBbqISIOwWs0uNLMB4LUyP30eE65CrSOqrTyqrTyqrTxBrm2Jc27S5b1qFuiVMLP1zrneWtcxGdVWHtVWHtVWnkatTS0XEZEGoUAXEWkQQQ3022tdwGGotvKotvKotvI0ZG2B7KGLiMihgjpCFxGRCRToIiINIlCBbma3mNlvzOw5M/uhmc0u2veXZtZnZi+a2ftqUNulZrbRzMbNrLdo+1IzS5nZM/7HUV+xYKra/H01fd0m1PI5M9tW9FpdUMt6/JrO91+bPjNbW+t6ipnZq2b2a/+1Wl/jWr5hZjvN7PmibceY2c/NbIv/75FZpqe82urivWZmi8zsfjN7wf8d/ZS/vbzXzjkXmA/gvUDEf/x54PP+4x7gWaAFWAa8BISPcm0nAm8GHgB6i7YvBZ6v8es2VW01f90m1Pk54L/X+n1WVE/Yf02WAzH/teqpdV1F9b0KzKt1HX4t7wJWFr/XgS8Aa/3Hawu/r3VSW12814CFwEr/cQew2f+9LOu1C9QI3Tn37865nP/0MbzVk8BbpPpu51zGOfcK0Ie3uPXRrG2Tc+7Fo/k9S3WY2mr+utW5/QukO+fGgMIC6TKBc+5BvLUQiq0G7vAf3wFcdFSL8k1RW11wzu1wzj3lPx4GNuGt0VzWaxeoQJ/gKuCn/uPJFrLuOuoVTW2ZmT1tZv9pZmfXupgi9fi6Xeu31L5Rqz/Ri9Tj61PMAf9uZhv8BdjrzbHOuR3+498Cx9aymEnU03sNM1sKnAY8Tpmv3VFdJLoUZvYfwO9Msuuzzrl7/GM+C+SAb9dbbZPYASx2zu02s9OBH5nZSc65oTqo7ag7XJ3AV4G/wQuqvwH+Du9/3DK5s5xz28xsAfBzM/uNPxqtO845Z2b1NEe6rt5rZtYO/D/gvznnhsxs/76ZvHZ1F+jOufMOt9/MPg58EDjX+Q0mSlvI+ojXNsXnZICM/3iDmb0EvAmo6kmscmrjKL1uxUqt08z+EfjxkaylBEf99ZkJ59w2/9+dZvZDvBZRPQX6G2a20Dm3w8wWAjtrXVCBc+6NwuNav9fMLIoX5t92zv2rv7ms1y5QLRczOx/4DHChcy5ZtGsdsMbMWsxsGbACeKIWNU5kZvPNLOw/Xo5X28u1rWq/unrd/DduwcXA81Mde5SUskB6TZhZm5l1FB7jTRio9es1UfHi8VcC9fSXYl2818wbin8d2OSc+/uiXeW9drU+yzvDM8J9eD3NZ/yPrxXt+yzejIQXgffXoLaL8XqsGeAN4D5/+x8AG/16nwI+VC+11cPrNqHOO4FfA8/5b+iFdfCeuwBv5sFLeO2rmtZTVNdyvFk3z/rvr5rWBnwHr72Y9d9rnwDmAr8AtgD/ARxTR7XVxXsNOAuv7fNcUa5dUO5rp0v/RUQaRKBaLiIiMjUFuohIg1Cgi4g0CAW6iEiDUKCLiDQIBbqISINQoIuINAgFuojPzM7wb9YU96/E3GhmJ9e6LpFS6cIikSJmdhMQBxJAv3Pu5hqXJFIyBbpIEf+eLU8CaeAdzrl8jUsSKZlaLiIHmwu0460eE69xLSIzohG6SBEzW4e3MtEyvBs2XVvjkkRKVnf3QxepFTO7Asg65+7yb3n8qJn9nnPul7WuTaQUGqGLiDQI9dBFRBqEAl1EpEEo0EVEGoQCXUSkQSjQRUQahAJdRKRBKNBFRBrE/wevWu83eVFLkgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/explore/miniconda3/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/loss.py:446: UserWarning: Using a target size (torch.Size([])) that is different to the input size (torch.Size([1])). This will likely lead to incorrect results due to broadcasting. Please ensure they have the same size.\n",
      "  return F.mse_loss(input, target, reduction=self.reduction)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/explore/miniconda3/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/loss.py:446: UserWarning: Using a target size (torch.Size([])) that is different to the input size (torch.Size([1])). This will likely lead to incorrect results due to broadcasting. Please ensure they have the same size.\n",
      "  return F.mse_loss(input, target, reduction=self.reduction)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'MSE loss')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Practice: Train the model with MSE Loss Function\n",
    "\n",
    "# Type your code here\n",
    "# Train the model\n",
    "# size of input \n",
    "D_in = 1\n",
    "# size of hidden layer \n",
    "H = 2\n",
    "# number of outputs \n",
    "D_out = 1\n",
    "# learning rate \n",
    "learning_rate = 0.1\n",
    "# create the model \n",
    "model2 = Net(D_in, H, D_out)\n",
    "#optimizer \n",
    "optimizer = torch.optim.SGD(model2.parameters(), lr=learning_rate)\n",
    "\n",
    "\n",
    "criterion_MSE = nn.MSELoss()\n",
    "cost_MSE = train(Y, X, model2, optimizer, criterion_MSE, epochs=1000)\n",
    "plt.plot(cost_MSE)\n",
    "plt.xlabel('epoch')\n",
    "plt.title('MSE loss')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Double-click <b>here</b> for the solution.\n",
    "\n",
    "<!-- \n",
    "learning_rate = 0.1\n",
    "criterion_mse=nn.MSELoss()\n",
    "model=Net(D_in,H,D_out)\n",
    "optimizer=torch.optim.SGD(model.parameters(), lr=learning_rate)\n",
    "cost_mse=train(Y,X,model,optimizer,criterion_mse,epochs=1000)\n",
    "plt.plot(cost_mse)\n",
    "plt.xlabel('epoch')\n",
    "plt.title('MSE loss ')\n",
    "-->\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a href=\"https://dataplatform.cloud.ibm.com/registration/stepone?context=cpdaas&apps=data_science_experience,watson_machine_learning\"><img src=\"https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBMDeveloperSkillsNetwork-DL0110EN-SkillsNetwork/Template/module%201/images/Watson_Studio.png\"/></a>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<!--Empty Space for separating topics-->\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2>About the Authors:</h2> \n",
    "\n",
    "<a href=\"https://www.linkedin.com/in/joseph-s-50398b136/\">Joseph Santarcangelo</a> has a PhD in Electrical Engineering, his research focused on using machine learning, signal processing, and computer vision to determine how videos impact human cognition. Joseph has been working for IBM since he completed his PhD. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Other contributors: <a href=\"https://www.linkedin.com/in/michelleccarey/\">Michelle Carey</a>, <a href=\"www.linkedin.com/in/jiahui-mavis-zhou-a4537814a\">Mavis Zhou</a>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Change Log\n",
    "\n",
    "| Date (YYYY-MM-DD) | Version | Changed By | Change Description                                          |\n",
    "| ----------------- | ------- | ---------- | ----------------------------------------------------------- |\n",
    "| 2020-09-23        | 2.0     | Shubham    | Migrated Lab to Markdown and added to course repo in GitLab |\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <h3 align=\"center\"> © IBM Corporation 2020. All rights reserved. <h3/>\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}